diff options
26 files changed, 455 insertions, 136 deletions
@@ -329,8 +329,8 @@ which means you can modify it, redistribute it or use it however you like. ## Subtitle Options: --write-sub Write subtitle file - --write-auto-sub Write automatic subtitle file (YouTube - only) + --write-auto-sub Write automatically generated subtitle file + (YouTube only) --all-subs Download all the available subtitles of the video --list-subs List all available subtitles for the video diff --git a/docs/supportedsites.md b/docs/supportedsites.md index 2e5283747..1df408610 100644 --- a/docs/supportedsites.md +++ b/docs/supportedsites.md @@ -129,6 +129,7 @@ - **Discovery** - **Dotsub** - **DouyuTV**: 斗鱼 + - **DPlay** - **dramafever** - **dramafever:series** - **DRBonanza** @@ -493,6 +494,7 @@ - **soompi:show** - **soundcloud** - **soundcloud:playlist** + - **soundcloud:search**: Soundcloud search - **soundcloud:set** - **soundcloud:user** - **soundgasm** @@ -706,6 +708,7 @@ - **youtube:show**: YouTube.com (multi-season) shows - **youtube:subscriptions**: YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication) - **youtube:user**: YouTube.com user videos (URL or "ytuser" keyword) + - **youtube:user:playlists**: YouTube.com user playlists - **youtube:watchlater**: Youtube watch later list, ":ytwatchlater" for short (requires authentication) - **Zapiks** - **ZDF** diff --git a/test/test_utils.py b/test/test_utils.py index ea1ff0547..501355c74 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -21,6 +21,7 @@ from youtube_dl.utils import ( clean_html, DateRange, detect_exe_version, + determine_ext, encodeFilename, escape_rfc3986, escape_url, @@ -238,6 +239,13 @@ class TestUtil(unittest.TestCase): self.assertEqual(unified_strdate('25-09-2014'), '20140925') self.assertEqual(unified_strdate('UNKNOWN DATE FORMAT'), None) + def test_determine_ext(self): + self.assertEqual(determine_ext('http://example.com/foo/bar.mp4/?download'), 'mp4') + self.assertEqual(determine_ext('http://example.com/foo/bar/?download', None), None) + self.assertEqual(determine_ext('http://example.com/foo/bar.nonext/?download', None), None) + self.assertEqual(determine_ext('http://example.com/foo/bar/mp4?download', None), None) + self.assertEqual(determine_ext('http://example.com/foo/bar.m3u8//?download'), 'm3u8') + def test_find_xpath_attr(self): testxml = '''<root> <node/> diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 1783ce01b..fba99af8d 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -156,7 +156,7 @@ class YoutubeDL(object): writethumbnail: Write the thumbnail image to a file write_all_thumbnails: Write all thumbnail formats to files writesubtitles: Write the video subtitles to a file - writeautomaticsub: Write the automatic subtitles to a file + writeautomaticsub: Write the automatically generated subtitles to a file allsubtitles: Downloads all the subtitles of the video (requires writesubtitles or writeautomaticsub) listsubtitles: Lists all available subtitles for the video @@ -833,6 +833,7 @@ class YoutubeDL(object): extra_info=extra) playlist_results.append(entry_result) ie_result['entries'] = playlist_results + self.to_screen('[download] Finished downloading playlist: %s' % playlist) return ie_result elif result_type == 'compat_list': self.report_warning( @@ -937,7 +938,7 @@ class YoutubeDL(object): filter_parts.append(string) def _remove_unused_ops(tokens): - # Remove operators that we don't use and join them with the sourrounding strings + # Remove operators that we don't use and join them with the surrounding strings # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' ALLOWED_OPS = ('/', '+', ',', '(', ')') last_string, last_start, last_end, last_line = None, None, None, None diff --git a/youtube_dl/downloader/common.py b/youtube_dl/downloader/common.py index 29a4500d3..b8bf8daf8 100644 --- a/youtube_dl/downloader/common.py +++ b/youtube_dl/downloader/common.py @@ -42,7 +42,7 @@ class FileDownloader(object): min_filesize: Skip files smaller than this size max_filesize: Skip files larger than this size xattr_set_filesize: Set ytdl.filesize user xattribute with expected size. - (experimenatal) + (experimental) external_downloader_args: A list of additional command-line arguments for the external downloader. diff --git a/youtube_dl/downloader/rtmp.py b/youtube_dl/downloader/rtmp.py index f1d219ba9..14d56db47 100644 --- a/youtube_dl/downloader/rtmp.py +++ b/youtube_dl/downloader/rtmp.py @@ -117,7 +117,7 @@ class RtmpFD(FileDownloader): return False # Download using rtmpdump. rtmpdump returns exit code 2 when - # the connection was interrumpted and resuming appears to be + # the connection was interrupted and resuming appears to be # possible. This is part of rtmpdump's normal usage, AFAIK. basic_args = [ 'rtmpdump', '--verbose', '-r', url, diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 59c82f65d..947b83683 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -132,6 +132,7 @@ from .dfb import DFBIE from .dhm import DHMIE from .dotsub import DotsubIE from .douyutv import DouyuTVIE +from .dplay import DPlayIE from .dramafever import ( DramaFeverIE, DramaFeverSeriesIE, @@ -575,7 +576,8 @@ from .soundcloud import ( SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE, - SoundcloudPlaylistIE + SoundcloudPlaylistIE, + SoundcloudSearchIE ) from .soundgasm import ( SoundgasmIE, @@ -832,6 +834,7 @@ from .youtube import ( YoutubeTruncatedIDIE, YoutubeTruncatedURLIE, YoutubeUserIE, + YoutubeUserPlaylistsIE, YoutubeWatchLaterIE, ) from .zapiks import ZapiksIE diff --git a/youtube_dl/extractor/bloomberg.py b/youtube_dl/extractor/bloomberg.py index 0dca29b71..11ace91dd 100644 --- a/youtube_dl/extractor/bloomberg.py +++ b/youtube_dl/extractor/bloomberg.py @@ -6,9 +6,9 @@ from .common import InfoExtractor class BloombergIE(InfoExtractor): - _VALID_URL = r'https?://www\.bloomberg\.com/news/videos/[^/]+/(?P<id>[^/?#]+)' + _VALID_URL = r'https?://www\.bloomberg\.com/news/[^/]+/[^/]+/(?P<id>[^/?#]+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.bloomberg.com/news/videos/b/aaeae121-5949-481e-a1ce-4562db6f5df2', # The md5 checksum changes 'info_dict': { @@ -17,7 +17,10 @@ class BloombergIE(InfoExtractor): 'title': 'Shah\'s Presentation on Foreign-Exchange Strategies', 'description': 'md5:a8ba0302912d03d246979735c17d2761', }, - } + }, { + 'url': 'http://www.bloomberg.com/news/articles/2015-11-12/five-strange-things-that-have-been-happening-in-financial-markets', + 'only_matching': True, + }] def _real_extract(self, url): name = self._match_id(url) diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 5e263f8b5..71bdcad5a 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -891,6 +891,11 @@ class InfoExtractor(object): if not media_nodes: manifest_version = '2.0' media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') + base_url = xpath_text( + manifest, ['{http://ns.adobe.com/f4m/1.0}baseURL', '{http://ns.adobe.com/f4m/2.0}baseURL'], + 'base URL', default=None) + if base_url: + base_url = base_url.strip() for i, media_el in enumerate(media_nodes): if manifest_version == '2.0': media_url = media_el.attrib.get('href') or media_el.attrib.get('url') @@ -898,7 +903,7 @@ class InfoExtractor(object): continue manifest_url = ( media_url if media_url.startswith('http://') or media_url.startswith('https://') - else ('/'.join(manifest_url.split('/')[:-1]) + '/' + media_url)) + else ((base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url)) # If media_url is itself a f4m manifest do the recursive extraction # since bitrates in parent manifest (this one) and media_url manifest # may differ leading to inability to resolve the format by requested diff --git a/youtube_dl/extractor/dplay.py b/youtube_dl/extractor/dplay.py new file mode 100644 index 000000000..6cda56a7f --- /dev/null +++ b/youtube_dl/extractor/dplay.py @@ -0,0 +1,51 @@ +# encoding: utf-8 +from __future__ import unicode_literals + +import time + +from .common import InfoExtractor +from ..utils import int_or_none + + +class DPlayIE(InfoExtractor): + _VALID_URL = r'http://www\.dplay\.se/[^/]+/(?P<id>[^/?#]+)' + + _TEST = { + 'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/', + 'info_dict': { + 'id': '3172', + 'ext': 'mp4', + 'display_id': 'season-1-svensken-lar-sig-njuta-av-livet', + 'title': 'Svensken lär sig njuta av livet', + 'duration': 2650, + }, + } + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex( + r'data-video-id="(\d+)"', webpage, 'video id') + + info = self._download_json( + 'http://www.dplay.se/api/v2/ajax/videos?video_id=' + video_id, + video_id)['data'][0] + + self._set_cookie( + 'secure.dplay.se', 'dsc-geo', + '{"countryCode":"NL","expiry":%d}' % ((time.time() + 20 * 60) * 1000)) + # TODO: consider adding support for 'stream_type=hds', it seems to + # require setting some cookies + manifest_url = self._download_json( + 'https://secure.dplay.se/secure/api/v2/user/authorization/stream/%s?stream_type=hls' % video_id, + video_id, 'Getting manifest url for hls stream')['hls'] + formats = self._extract_m3u8_formats( + manifest_url, video_id, ext='mp4', entry_protocol='m3u8_native') + + return { + 'id': video_id, + 'display_id': display_id, + 'title': info['title'], + 'formats': formats, + 'duration': int_or_none(info.get('video_metadata_length'), scale=1000), + } diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 51516a38a..2b934148d 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -823,6 +823,19 @@ class GenericIE(InfoExtractor): 'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014', }, }, + # Kaltura embed protected with referrer + { + 'url': 'http://www.disney.nl/disney-channel/filmpjes/achter-de-schermen#/videoId/violetta-achter-de-schermen-ruggero', + 'info_dict': { + 'id': '1_g4fbemnq', + 'ext': 'mp4', + 'title': 'Violetta - Achter De Schermen - Ruggero', + 'description': 'Achter de schermen met Ruggero', + 'timestamp': 1435133761, + 'upload_date': '20150624', + 'uploader_id': 'echojecka', + }, + }, # Eagle.Platform embed (generic URL) { 'url': 'http://lenta.ru/news/2015/03/06/navalny/', @@ -1045,6 +1058,20 @@ class GenericIE(InfoExtractor): 'description': 'Tabletop: Dread, Last Thoughts', 'duration': 51690, }, + }, + # JWPlayer with M3U8 + { + 'url': 'http://ren.tv/novosti/2015-09-25/sluchaynyy-prohozhiy-poymal-avtougonshchika-v-murmanske-video', + 'info_dict': { + 'id': 'playlist', + 'ext': 'mp4', + 'title': 'Случайный прохожий поймал автоугонщика в Мурманске. ВИДЕО | РЕН ТВ', + 'uploader': 'ren.tv', + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + } } ] @@ -1694,7 +1721,9 @@ class GenericIE(InfoExtractor): mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_?[Ii]d'\s*:\s*'(?P<id>[^']+)',", webpage) or re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage)) if mobj is not None: - return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura') + return self.url_result(smuggle_url( + 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), + {'source_url': url}), 'Kaltura') # Look for Eagle.Platform embeds mobj = re.search( @@ -1739,7 +1768,7 @@ class GenericIE(InfoExtractor): # Look for UDN embeds mobj = re.search( - r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage) + r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._PROTOCOL_RELATIVE_VALID_URL, webpage) if mobj is not None: return self.url_result( compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed') @@ -1859,6 +1888,7 @@ class GenericIE(InfoExtractor): entries = [] for video_url in found: + video_url = video_url.replace('\\/', '/') video_url = compat_urlparse.urljoin(url, video_url) video_id = compat_urllib_parse_unquote(os.path.basename(video_url)) @@ -1870,25 +1900,24 @@ class GenericIE(InfoExtractor): # here's a fun little line of code for you: video_id = os.path.splitext(video_id)[0] + entry_info_dict = { + 'id': video_id, + 'uploader': video_uploader, + 'title': video_title, + 'age_limit': age_limit, + } + ext = determine_ext(video_url) if ext == 'smil': - entries.append({ - 'id': video_id, - 'formats': self._extract_smil_formats(video_url, video_id), - 'uploader': video_uploader, - 'title': video_title, - 'age_limit': age_limit, - }) + entry_info_dict['formats'] = self._extract_smil_formats(video_url, video_id) elif ext == 'xspf': return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id) + elif ext == 'm3u8': + entry_info_dict['formats'] = self._extract_m3u8_formats(video_url, video_id, ext='mp4') else: - entries.append({ - 'id': video_id, - 'url': video_url, - 'uploader': video_uploader, - 'title': video_title, - 'age_limit': age_limit, - }) + entry_info_dict['url'] = video_url + + entries.append(entry_info_dict) if len(entries) == 1: return entries[0] diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py index fce179000..c158f2064 100644 --- a/youtube_dl/extractor/instagram.py +++ b/youtube_dl/extractor/instagram.py @@ -10,7 +10,7 @@ from ..utils import ( class InstagramIE(InfoExtractor): - _VALID_URL = r'https://instagram\.com/p/(?P<id>[^/?#&]+)' + _VALID_URL = r'https?://(?:www\.)?instagram\.com/p/(?P<id>[^/?#&]+)' _TESTS = [{ 'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc', 'md5': '0d2da106a9d2631273e192b372806516', diff --git a/youtube_dl/extractor/kaltura.py b/youtube_dl/extractor/kaltura.py index 0dcd6cd05..583b1a5ad 100644 --- a/youtube_dl/extractor/kaltura.py +++ b/youtube_dl/extractor/kaltura.py @@ -2,12 +2,18 @@ from __future__ import unicode_literals import re +import base64 from .common import InfoExtractor -from ..compat import compat_urllib_parse +from ..compat import ( + compat_urllib_parse, + compat_urlparse, +) from ..utils import ( + clean_html, ExtractorError, int_or_none, + unsmuggle_url, ) @@ -121,31 +127,47 @@ class KalturaIE(InfoExtractor): video_id, actions, note='Downloading video info JSON') def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + mobj = re.match(self._VALID_URL, url) partner_id = mobj.group('partner_id_s') or mobj.group('partner_id') or mobj.group('partner_id_html5') entry_id = mobj.group('id_s') or mobj.group('id') or mobj.group('id_html5') info, source_data = self._get_video_info(entry_id, partner_id) - formats = [{ - 'format_id': '%(fileExt)s-%(bitrate)s' % f, - 'ext': f['fileExt'], - 'tbr': f['bitrate'], - 'fps': f.get('frameRate'), - 'filesize_approx': int_or_none(f.get('size'), invscale=1024), - 'container': f.get('containerFormat'), - 'vcodec': f.get('videoCodecId'), - 'height': f.get('height'), - 'width': f.get('width'), - 'url': '%s/flavorId/%s' % (info['dataUrl'], f['id']), - } for f in source_data['flavorAssets']] + source_url = smuggled_data.get('source_url') + if source_url: + referrer = base64.b64encode( + '://'.join(compat_urlparse.urlparse(source_url)[:2]) + .encode('utf-8')).decode('utf-8') + else: + referrer = None + + formats = [] + for f in source_data['flavorAssets']: + video_url = '%s/flavorId/%s' % (info['dataUrl'], f['id']) + if referrer: + video_url += '?referrer=%s' % referrer + formats.append({ + 'format_id': '%(fileExt)s-%(bitrate)s' % f, + 'ext': f.get('fileExt'), + 'tbr': int_or_none(f['bitrate']), + 'fps': int_or_none(f.get('frameRate')), + 'filesize_approx': int_or_none(f.get('size'), invscale=1024), + 'container': f.get('containerFormat'), + 'vcodec': f.get('videoCodecId'), + 'height': int_or_none(f.get('height')), + 'width': int_or_none(f.get('width')), + 'url': video_url, + }) + self._check_formats(formats, entry_id) self._sort_formats(formats) return { 'id': entry_id, 'title': info['name'], 'formats': formats, - 'description': info.get('description'), + 'description': clean_html(info.get('description')), 'thumbnail': info.get('thumbnailUrl'), 'duration': info.get('duration'), 'timestamp': info.get('createdAt'), diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index 8fb9b1849..b787e2a73 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -22,7 +22,7 @@ class PBSIE(InfoExtractor): # Article with embedded player (or direct video) (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+?)(?:\.html)?/?(?:$|[?\#]) | # Player - video\.pbs\.org/(?:widget/)?partnerplayer/(?P<player_id>[^/]+)/ + (?:video|player)\.pbs\.org/(?:widget/)?partnerplayer/(?P<player_id>[^/]+)/ ) ''' @@ -170,6 +170,10 @@ class PBSIE(InfoExtractor): 'params': { 'skip_download': True, # requires ffmpeg }, + }, + { + 'url': 'http://player.pbs.org/widget/partnerplayer/2365297708/?start=0&end=0&chapterbar=false&endscreen=false&topbar=true', + 'only_matching': True, } ] _ERRORS = { @@ -259,7 +263,7 @@ class PBSIE(InfoExtractor): return self.playlist_result(entries, display_id) info = self._download_json( - 'http://video.pbs.org/videoInfo/%s?format=json&type=partner' % video_id, + 'http://player.pbs.org/videoInfo/%s?format=json&type=partner' % video_id, display_id) formats = [] diff --git a/youtube_dl/extractor/pluralsight.py b/youtube_dl/extractor/pluralsight.py index fd32836cc..792316db8 100644 --- a/youtube_dl/extractor/pluralsight.py +++ b/youtube_dl/extractor/pluralsight.py @@ -1,7 +1,8 @@ from __future__ import unicode_literals -import re import json +import random +import collections from .common import InfoExtractor from ..compat import ( @@ -17,13 +18,18 @@ from ..utils import ( ) -class PluralsightIE(InfoExtractor): +class PluralsightBaseIE(InfoExtractor): + _API_BASE = 'http://app.pluralsight.com' + + +class PluralsightIE(PluralsightBaseIE): IE_NAME = 'pluralsight' - _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/training/player\?author=(?P<author>[^&]+)&name=(?P<name>[^&]+)(?:&mode=live)?&clip=(?P<clip>\d+)&course=(?P<course>[^&]+)' - _LOGIN_URL = 'https://www.pluralsight.com/id/' + _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/training/player\?' + _LOGIN_URL = 'https://app.pluralsight.com/id/' + _NETRC_MACHINE = 'pluralsight' - _TEST = { + _TESTS = [{ 'url': 'http://www.pluralsight.com/training/player?author=mike-mckeown&name=hosting-sql-server-windows-azure-iaas-m7-mgmt&mode=live&clip=3&course=hosting-sql-server-windows-azure-iaas', 'md5': '4d458cf5cf4c593788672419a8dd4cf8', 'info_dict': { @@ -33,7 +39,14 @@ class PluralsightIE(InfoExtractor): 'duration': 338, }, 'skip': 'Requires pluralsight account credentials', - } + }, { + 'url': 'https://app.pluralsight.com/training/player?course=angularjs-get-started&author=scott-allen&name=angularjs-get-started-m1-introduction&clip=0&mode=live', + 'only_matching': True, + }, { + # available without pluralsight account + 'url': 'http://app.pluralsight.com/training/player?author=scott-allen&name=angularjs-get-started-m1-introduction&mode=live&clip=0&course=angularjs-get-started', + 'only_matching': True, + }] def _real_initialize(self): self._login() @@ -41,7 +54,7 @@ class PluralsightIE(InfoExtractor): def _login(self): (username, password) = self._get_login_info() if username is None: - self.raise_login_required('Pluralsight account is required') + return login_page = self._download_webpage( self._LOGIN_URL, None, 'Downloading login page') @@ -73,31 +86,48 @@ class PluralsightIE(InfoExtractor): if error: raise ExtractorError('Unable to login: %s' % error, expected=True) + if all(p not in response for p in ('__INITIAL_STATE__', '"currentUser"')): + raise ExtractorError('Unable to log in') + def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - author = mobj.group('author') - name = mobj.group('name') - clip_id = mobj.group('clip') - course = mobj.group('course') + qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + + author = qs.get('author', [None])[0] + name = qs.get('name', [None])[0] + clip_id = qs.get('clip', [None])[0] + course = qs.get('course', [None])[0] + + if any(not f for f in (author, name, clip_id, course,)): + raise ExtractorError('Invalid URL', expected=True) display_id = '%s-%s' % (name, clip_id) webpage = self._download_webpage(url, display_id) - collection = self._parse_json( - self._search_regex( - r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)', - webpage, 'modules'), - display_id) + modules = self._search_regex( + r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)', + webpage, 'modules', default=None) + + if modules: + collection = self._parse_json(modules, display_id) + else: + # Webpage may be served in different layout (see + # https://github.com/rg3/youtube-dl/issues/7607) + collection = self._parse_json( + self._search_regex( + r'var\s+initialState\s*=\s*({.+?});\n', webpage, 'initial state'), + display_id)['course']['modules'] module, clip = None, None for module_ in collection: - if module_.get('moduleName') == name: + if name in (module_.get('moduleName'), module_.get('name')): module = module_ for clip_ in module_.get('clips', []): clip_index = clip_.get('clipIndex') if clip_index is None: + clip_index = clip_.get('index') + if clip_index is None: continue if compat_str(clip_index) == clip_id: clip = clip_ @@ -112,13 +142,33 @@ class PluralsightIE(InfoExtractor): 'high': {'width': 1024, 'height': 768}, } + AllowedQuality = collections.namedtuple('AllowedQuality', ['ext', 'qualities']) + ALLOWED_QUALITIES = ( - ('webm', ('high',)), - ('mp4', ('low', 'medium', 'high',)), + AllowedQuality('webm', ('high',)), + AllowedQuality('mp4', ('low', 'medium', 'high',)), ) + # In order to minimize the number of calls to ViewClip API and reduce + # the probability of being throttled or banned by Pluralsight we will request + # only single format until formats listing was explicitly requested. + if self._downloader.params.get('listformats', False): + allowed_qualities = ALLOWED_QUALITIES + else: + def guess_allowed_qualities(): + req_format = self._downloader.params.get('format') or 'best' + req_format_split = req_format.split('-') + if len(req_format_split) > 1: + req_ext, req_quality = req_format_split + for allowed_quality in ALLOWED_QUALITIES: + if req_ext == allowed_quality.ext and req_quality in allowed_quality.qualities: + return (AllowedQuality(req_ext, (req_quality, )), ) + req_ext = 'webm' if self._downloader.params.get('prefer_free_formats') else 'mp4' + return (AllowedQuality(req_ext, ('high', )), ) + allowed_qualities = guess_allowed_qualities() + formats = [] - for ext, qualities in ALLOWED_QUALITIES: + for ext, qualities in allowed_qualities: for quality in qualities: f = QUALITIES[quality].copy() clip_post = { @@ -132,12 +182,23 @@ class PluralsightIE(InfoExtractor): 'q': '%dx%d' % (f['width'], f['height']), } request = compat_urllib_request.Request( - 'http://www.pluralsight.com/training/Player/ViewClip', + '%s/training/Player/ViewClip' % self._API_BASE, json.dumps(clip_post).encode('utf-8')) request.add_header('Content-Type', 'application/json;charset=utf-8') format_id = '%s-%s' % (ext, quality) clip_url = self._download_webpage( request, display_id, 'Downloading %s URL' % format_id, fatal=False) + + # Pluralsight tracks multiple sequential calls to ViewClip API and start + # to return 429 HTTP errors after some time (see + # https://github.com/rg3/youtube-dl/pull/6989). Moreover it may even lead + # to account ban (see https://github.com/rg3/youtube-dl/issues/6842). + # To somewhat reduce the probability of these consequences + # we will sleep random amount of time before each call to ViewClip. + self._sleep( + random.randint(2, 5), display_id, + '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling') + if not clip_url: continue f.update({ @@ -163,10 +224,10 @@ class PluralsightIE(InfoExtractor): } -class PluralsightCourseIE(InfoExtractor): +class PluralsightCourseIE(PluralsightBaseIE): IE_NAME = 'pluralsight:course' - _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/courses/(?P<id>[^/]+)' - _TEST = { + _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:library/)?courses/(?P<id>[^/]+)' + _TESTS = [{ # Free course from Pluralsight Starter Subscription for Microsoft TechNet # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz 'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas', @@ -176,7 +237,14 @@ class PluralsightCourseIE(InfoExtractor): 'description': 'md5:61b37e60f21c4b2f91dc621a977d0986', }, 'playlist_count': 31, - } + }, { + # available without pluralsight account + 'url': 'https://www.pluralsight.com/courses/angularjs-get-started', + 'only_matching': True, + }, { + 'url': 'https://app.pluralsight.com/library/courses/understanding-microsoft-azure-amazon-aws/table-of-contents', + 'only_matching': True, + }] def _real_extract(self, url): course_id = self._match_id(url) @@ -184,14 +252,14 @@ class PluralsightCourseIE(InfoExtractor): # TODO: PSM cookie course = self._download_json( - 'http://www.pluralsight.com/data/course/%s' % course_id, + '%s/data/course/%s' % (self._API_BASE, course_id), course_id, 'Downloading course JSON') title = course['title'] description = course.get('description') or course.get('shortDescription') course_data = self._download_json( - 'http://www.pluralsight.com/data/course/content/%s' % course_id, + '%s/data/course/content/%s' % (self._API_BASE, course_id), course_id, 'Downloading course data JSON') entries = [] @@ -201,7 +269,7 @@ class PluralsightCourseIE(InfoExtractor): if not player_parameters: continue entries.append(self.url_result( - 'http://www.pluralsight.com/training/player?%s' % player_parameters, + '%s/training/player?%s' % (self._API_BASE, player_parameters), 'Pluralsight')) return self.playlist_result(entries, course_id, title, description) diff --git a/youtube_dl/extractor/pornhd.py b/youtube_dl/extractor/pornhd.py index dbb2c3bd9..57c78ba52 100644 --- a/youtube_dl/extractor/pornhd.py +++ b/youtube_dl/extractor/pornhd.py @@ -36,7 +36,8 @@ class PornHdIE(InfoExtractor): webpage = self._download_webpage(url, display_id or video_id) title = self._html_search_regex( - r'<title>(.+) porn HD.+?</title>', webpage, 'title') + [r'<span[^>]+class=["\']video-name["\'][^>]*>([^<]+)', + r'<title>(.+?) - .*?[Pp]ornHD.*?</title>'], webpage, 'title') description = self._html_search_regex( r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False) view_count = int_or_none(self._html_search_regex( diff --git a/youtube_dl/extractor/rtve.py b/youtube_dl/extractor/rtve.py index 5b97d33ca..0fe6356db 100644 --- a/youtube_dl/extractor/rtve.py +++ b/youtube_dl/extractor/rtve.py @@ -6,7 +6,7 @@ import re import time from .common import InfoExtractor -from ..compat import compat_urllib_request, compat_urlparse +from ..compat import compat_urllib_request from ..utils import ( ExtractorError, float_or_none, @@ -107,15 +107,9 @@ class RTVEALaCartaIE(InfoExtractor): png = self._download_webpage(png_request, video_id, 'Downloading url information') video_url = _decrypt_url(png) if not video_url.endswith('.f4m'): - auth_url = video_url.replace( + video_url = video_url.replace( 'resources/', 'auth/resources/' ).replace('.net.rtve', '.multimedia.cdn.rtve') - video_path = self._download_webpage( - auth_url, video_id, 'Getting video url') - # Use mvod1.akcdn instead of flash.akamaihd.multimedia.cdn to get - # the right Content-Length header and the mp4 format - video_url = compat_urlparse.urljoin( - 'http://mvod1.akcdn.rtve.es/', video_path) subtitles = None if info.get('sbtFile') is not None: diff --git a/youtube_dl/extractor/rutube.py b/youtube_dl/extractor/rutube.py index d94dc7399..6b09550b0 100644 --- a/youtube_dl/extractor/rutube.py +++ b/youtube_dl/extractor/rutube.py @@ -9,7 +9,7 @@ from ..compat import ( compat_str, ) from ..utils import ( - ExtractorError, + determine_ext, unified_strdate, ) @@ -51,10 +51,25 @@ class RutubeIE(InfoExtractor): 'http://rutube.ru/api/play/options/%s/?format=json' % video_id, video_id, 'Downloading options JSON') - m3u8_url = options['video_balancer'].get('m3u8') - if m3u8_url is None: - raise ExtractorError('Couldn\'t find m3u8 manifest url') - formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4') + formats = [] + for format_id, format_url in options['video_balancer'].items(): + ext = determine_ext(format_url) + if ext == 'm3u8': + m3u8_formats = self._extract_m3u8_formats( + format_url, video_id, 'mp4', m3u8_id=format_id, fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + elif ext == 'f4m': + f4m_formats = self._extract_f4m_formats( + format_url, video_id, f4m_id=format_id, fatal=False) + if f4m_formats: + formats.extend(f4m_formats) + else: + formats.append({ + 'url': format_url, + 'format_id': format_id, + }) + self._sort_formats(formats) return { 'id': video['id'], @@ -74,9 +89,9 @@ class RutubeIE(InfoExtractor): class RutubeEmbedIE(InfoExtractor): IE_NAME = 'rutube:embed' IE_DESC = 'Rutube embedded videos' - _VALID_URL = 'https?://rutube\.ru/video/embed/(?P<id>[0-9]+)' + _VALID_URL = 'https?://rutube\.ru/(?:video|play)/embed/(?P<id>[0-9]+)' - _TEST = { + _TESTS = [{ 'url': 'http://rutube.ru/video/embed/6722881?vk_puid37=&vk_puid38=', 'info_dict': { 'id': 'a10e53b86e8f349080f718582ce4c661', @@ -90,7 +105,10 @@ class RutubeEmbedIE(InfoExtractor): 'params': { 'skip_download': 'Requires ffmpeg', }, - } + }, { + 'url': 'http://rutube.ru/play/embed/8083783', + 'only_matching': True, + }] def _real_extract(self, url): embed_id = self._match_id(url) diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index 2b60d354a..02e64e094 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -4,13 +4,17 @@ from __future__ import unicode_literals import re import itertools -from .common import InfoExtractor +from .common import ( + InfoExtractor, + SearchInfoExtractor +) from ..compat import ( compat_str, compat_urlparse, compat_urllib_parse, ) from ..utils import ( + encode_dict, ExtractorError, int_or_none, unified_strdate, @@ -469,3 +473,60 @@ class SoundcloudPlaylistIE(SoundcloudIE): 'description': data.get('description'), 'entries': entries, } + + +class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE): + IE_NAME = 'soundcloud:search' + IE_DESC = 'Soundcloud search' + _MAX_RESULTS = float('inf') + _TESTS = [{ + 'url': 'scsearch15:post-avant jazzcore', + 'info_dict': { + 'title': 'post-avant jazzcore', + }, + 'playlist_count': 15, + }] + + _SEARCH_KEY = 'scsearch' + _MAX_RESULTS_PER_PAGE = 200 + _DEFAULT_RESULTS_PER_PAGE = 50 + _API_V2_BASE = 'https://api-v2.soundcloud.com' + + def _get_collection(self, endpoint, collection_id, **query): + limit = min( + query.get('limit', self._DEFAULT_RESULTS_PER_PAGE), + self._MAX_RESULTS_PER_PAGE) + query['limit'] = limit + query['client_id'] = self._CLIENT_ID + query['linked_partitioning'] = '1' + query['offset'] = 0 + data = compat_urllib_parse.urlencode(encode_dict(query)) + next_url = '{0}{1}?{2}'.format(self._API_V2_BASE, endpoint, data) + + collected_results = 0 + + for i in itertools.count(1): + response = self._download_json( + next_url, collection_id, 'Downloading page {0}'.format(i), + 'Unable to download API page') + + collection = response.get('collection', []) + if not collection: + break + + collection = list(filter(bool, collection)) + collected_results += len(collection) + + for item in collection: + yield self.url_result(item['uri'], SoundcloudIE.ie_key()) + + if not collection or collected_results >= limit: + break + + next_url = response.get('next_href') + if not next_url: + break + + def _get_n_results(self, query, n): + tracks = self._get_collection('/search/tracks', query, limit=n, q=query) + return self.playlist_result(tracks, playlist_title=query) diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 43315e75d..1555aa77c 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -187,8 +187,12 @@ class ThePlatformIE(ThePlatformBaseIE): # Seems there's no pattern for the interested script filename, so # I try one by one for script in reversed(scripts): - feed_script = self._download_webpage(script, video_id, 'Downloading feed script') - feed_id = self._search_regex(r'defaultFeedId\s*:\s*"([^"]+)"', feed_script, 'default feed id', default=None) + feed_script = self._download_webpage( + self._proto_relative_url(script, 'http:'), + video_id, 'Downloading feed script') + feed_id = self._search_regex( + r'defaultFeedId\s*:\s*"([^"]+)"', feed_script, + 'default feed id', default=None) if feed_id is not None: break if feed_id is None: diff --git a/youtube_dl/extractor/udn.py b/youtube_dl/extractor/udn.py index 2151f8338..ee35b7227 100644 --- a/youtube_dl/extractor/udn.py +++ b/youtube_dl/extractor/udn.py @@ -12,7 +12,8 @@ from ..compat import compat_urlparse class UDNEmbedIE(InfoExtractor): IE_DESC = '聯合影音' - _VALID_URL = r'https?://video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)' + _PROTOCOL_RELATIVE_VALID_URL = r'//video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)' + _VALID_URL = r'https?:' + _PROTOCOL_RELATIVE_VALID_URL _TESTS = [{ 'url': 'http://video.udn.com/embed/news/300040', 'md5': 'de06b4c90b042c128395a88f0384817e', diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index b72341a2b..057c72f39 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -189,6 +189,10 @@ class VimeoIE(VimeoBaseInfoExtractor): 'note': 'Video not completely processed, "failed" seed status', 'only_matching': True, }, + { + 'url': 'https://vimeo.com/groups/travelhd/videos/22439234', + 'only_matching': True, + }, ] @staticmethod @@ -486,8 +490,7 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): password_request, list_id, 'Verifying the password', 'Wrong password') - def _extract_videos(self, list_id, base_url): - video_ids = [] + def _title_and_entries(self, list_id, base_url): for pagenum in itertools.count(1): page_url = self._page_url(base_url, pagenum) webpage = self._download_webpage( @@ -496,18 +499,18 @@ class VimeoChannelIE(VimeoBaseInfoExtractor): if pagenum == 1: webpage = self._login_list_password(page_url, list_id, webpage) + yield self._extract_list_title(webpage) + + for video_id in re.findall(r'id="clip_(\d+?)"', webpage): + yield self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo') - video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage)) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: break - entries = [self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo') - for video_id in video_ids] - return {'_type': 'playlist', - 'id': list_id, - 'title': self._extract_list_title(webpage), - 'entries': entries, - } + def _extract_videos(self, list_id, base_url): + title_and_entries = self._title_and_entries(list_id, base_url) + list_title = next(title_and_entries) + return self.playlist_result(title_and_entries, list_id, list_title) def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -568,7 +571,7 @@ class VimeoAlbumIE(VimeoChannelIE): class VimeoGroupsIE(VimeoAlbumIE): IE_NAME = 'vimeo:group' - _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)' + _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)' _TESTS = [{ 'url': 'https://vimeo.com/groups/rolexawards', 'info_dict': { diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 247769067..1580c54fe 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -178,15 +178,13 @@ class YoutubeBaseInfoExtractor(InfoExtractor): return -class YoutubePlaylistBaseInfoExtractor(InfoExtractor): - # Extract the video ids from the playlist pages +class YoutubeEntryListBaseInfoExtractor(InfoExtractor): + # Extract entries from page with "Load more" button def _entries(self, page, playlist_id): more_widget_html = content_html = page for page_num in itertools.count(1): - for video_id, video_title in self.extract_videos_from_page(content_html): - yield self.url_result( - video_id, 'Youtube', video_id=video_id, - video_title=video_title) + for entry in self._process_page(content_html): + yield entry mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) if not mobj: @@ -203,6 +201,12 @@ class YoutubePlaylistBaseInfoExtractor(InfoExtractor): break more_widget_html = more['load_more_widget_html'] + +class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): + def _process_page(self, content): + for video_id, video_title in self.extract_videos_from_page(content): + yield self.url_result(video_id, 'Youtube', video_id, video_title) + def extract_videos_from_page(self, page): ids_in_page = [] titles_in_page = [] @@ -224,6 +228,19 @@ class YoutubePlaylistBaseInfoExtractor(InfoExtractor): return zip(ids_in_page, titles_in_page) +class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): + def _process_page(self, content): + for playlist_id in re.findall(r'href="/?playlist\?list=(.+?)"', content): + yield self.url_result( + 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist') + + def _real_extract(self, url): + playlist_id = self._match_id(url) + webpage = self._download_webpage(url, playlist_id) + title = self._og_search_title(webpage, fatal=False) + return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title) + + class YoutubeIE(YoutubeBaseInfoExtractor): IE_DESC = 'YouTube.com' _VALID_URL = r"""(?x)^ @@ -1637,7 +1654,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtract self.report_warning('Youtube gives an alert message: ' + match) playlist_title = self._html_search_regex( - r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>', + r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>', page, 'title') return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title) @@ -1764,6 +1781,29 @@ class YoutubeUserIE(YoutubeChannelIE): return super(YoutubeUserIE, cls).suitable(url) +class YoutubeUserPlaylistsIE(YoutubePlaylistsBaseInfoExtractor): + IE_DESC = 'YouTube.com user playlists' + _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/user/(?P<id>[^/]+)/playlists' + IE_NAME = 'youtube:user:playlists' + + _TESTS = [{ + 'url': 'http://www.youtube.com/user/ThirstForScience/playlists', + 'playlist_mincount': 4, + 'info_dict': { + 'id': 'ThirstForScience', + 'title': 'Thirst for Science', + }, + }, { + # with "Load more" button + 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd', + 'playlist_mincount': 70, + 'info_dict': { + 'id': 'igorkle1', + 'title': 'Игорь Клейнер', + }, + }] + + class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE): IE_DESC = 'YouTube.com searches' # there doesn't appear to be a real limit, for example if you search for @@ -1859,7 +1899,7 @@ class YoutubeSearchURLIE(InfoExtractor): } -class YoutubeShowIE(InfoExtractor): +class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor): IE_DESC = 'YouTube.com (multi-season) shows' _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)' IE_NAME = 'youtube:show' @@ -1873,26 +1913,9 @@ class YoutubeShowIE(InfoExtractor): }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - playlist_id = mobj.group('id') - webpage = self._download_webpage( - 'https://www.youtube.com/show/%s/playlists' % playlist_id, playlist_id, 'Downloading show webpage') - # There's one playlist for each season of the show - m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) - self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons))) - entries = [ - self.url_result( - 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist') - for season in m_seasons - ] - title = self._og_search_title(webpage, fatal=False) - - return { - '_type': 'playlist', - 'id': playlist_id, - 'title': title, - 'entries': entries, - } + playlist_id = self._match_id(url) + return super(YoutubeShowIE, self)._real_extract( + 'https://www.youtube.com/show/%s/playlists' % playlist_id) class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): diff --git a/youtube_dl/options.py b/youtube_dl/options.py index 3dd6d290b..079fe7e8a 100644 --- a/youtube_dl/options.py +++ b/youtube_dl/options.py @@ -363,7 +363,7 @@ def parseOpts(overrideArguments=None): subtitles.add_option( '--write-auto-sub', '--write-automatic-sub', action='store_true', dest='writeautomaticsub', default=False, - help='Write automatic subtitle file (YouTube only)') + help='Write automatically generated subtitle file (YouTube only)') subtitles.add_option( '--all-subs', action='store_true', dest='allsubtitles', default=False, diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index d00b14b86..c0325f054 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -925,6 +925,21 @@ def determine_ext(url, default_ext='unknown_video'): guess = url.partition('?')[0].rpartition('.')[2] if re.match(r'^[A-Za-z0-9]+$', guess): return guess + elif guess.rstrip('/') in ( + 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac', + 'flv', 'f4v', 'f4a', 'f4b', + 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus', + 'mkv', 'mka', 'mk3d', + 'avi', 'divx', + 'mov', + 'asf', 'wmv', 'wma', + '3gp', '3g2', + 'mp3', + 'flac', + 'ape', + 'wav', + 'f4f', 'f4m', 'm3u8', 'smil'): + return guess.rstrip('/') else: return default_ext @@ -1668,7 +1683,9 @@ def urlencode_postdata(*args, **kargs): def encode_dict(d, encoding='utf-8'): - return dict((k.encode(encoding), v.encode(encoding)) for k, v in d.items()) + def encode(v): + return v.encode(encoding) if isinstance(v, compat_basestring) else v + return dict((encode(k), encode(v)) for k, v in d.items()) US_RATINGS = { diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 6f601cbb1..2baf1ac42 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2015.11.15' +__version__ = '2015.11.21' |